Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Improving Local Identifiability in Probabilistic Box Embeddings

About

Geometric embeddings have recently received attention for their natural ability to represent transitive asymmetric relations via containment. Box embeddings, where objects are represented by n-dimensional hyperrectangles, are a particularly promising example of such an embedding as they are closed under intersection and their volume can be calculated easily, allowing them to naturally represent calibrated probability distributions. The benefits of geometric embeddings also introduce a problem of local identifiability, however, where whole neighborhoods of parameters result in equivalent loss which impedes learning. Prior work addressed some of these issues by using an approximation to Gaussian convolution over the box parameters, however, this intersection operation also increases the sparsity of the gradient. In this work, we model the box parameters with min and max Gumbel distributions, which were chosen such that space is still closed under the operation of the intersection. The calculation of the expected intersection volume involves all parameters, and we demonstrate experimentally that this drastically improves the ability of such models to learn.

Shib Sankar Dasgupta, Michael Boratko, Dongxu Zhang, Luke Vilnis, Xiang Lorraine Li, Andrew McCallum• 2020

Related benchmarks

TaskDatasetResultRank
Taxonomy ExpansionScience single-parent hierarchies (test)
R@129.1
13
Taxonomy ExpansionEnvironment single-parent hierarchies (test)
R@133.6
13
Taxonomy ExpansionWordNet single-parent hierarchies (test)
R@121.6
13
Taxonomy ExpansionCaltech-UCSD Birds-200 2011 (test)
Precision@173.93
10
Showing 4 of 4 rows

Other info

Follow for update